Music Classification Based on Sequential Naive Bayes and Music Score Data

• Activity Number: 248 - Machine Learning in Science and Industry
• Type: Contributed
• Date/Time: Monday, July 29, 2019 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistical Learning and Data Science
• Author(s): Tunan Ren* and Hansheng Wang and Feifei Wang
• Companies: Guanghua School of Management and Guanghua School of Management, Peking University, Beijing, China and School of Statistics, Renmin University of China, Beijing, China
• Keywords: BIC; Music Classification; Music Score; Selection Consistency; Sequential Naive Bayes

Abstract

Due to the rapid development of digital music market, online music websites are widely available in our daily life.For these websites, one important objective is to develop automatic music classification algorithms to manage a huge amount of music.Among all available information sources for music classification, music score is believed to contain the most basic information of music. Particularly, the transition in pitches produces a melody, and the transition in beats produces a rhythm, both of which help decide the classes of music. Therefore, we focus on music score data and propose a sequential naive Bayes method for music classification. This method can be viewed as an novel extension of the classical naive Bayes classifier, but takes the transitional information between pitches and beats into consideration. To sufficiently reduce the number of estimated parameters, we propose a BIC-type criterion and develop a computationally efficient algorithm for model selection. The selection consistency of the BIC method is investigated.The finite sample performance is assessed through both simulations and a real music dataset.